Hi,
I am a newbie at R and was thinking of ways to vectorize this loop but can't
quite figure out if anything more can be done :
Here the dimensions of the arrays are (N=2000, K=2, D=1000):
responsibilities : N * K
pi: K
corp: N*D
theta: K*D
for (n in 1:N)
{
for (k in 1:K
On Sun, 9 Mar 2008, Gabor Grothendieck wrote:
> Try this:
>
> r <- unlist(sapply(x, function(x) c(rep(0, x), 1)))
And in large problems ( length(x) >> 1000 ), this will be faster still
r <- rep( rep( 0:1, length(x) ), rbind(x,1) )
HTH,
Chuck
> x2 <- diff(c(0, which(r==1))) - 1
> iden
Try this:
r <- unlist(sapply(x, function(x) c(rep(0, x), 1)))
x2 <- diff(c(0, which(r==1))) - 1
identical(x, x2)
On Sun, Mar 9, 2008 at 7:55 PM, remko duursma <[EMAIL PROTECTED]> wrote:
>
> Dear R-helpers,
>
> I have two problems that I don't know how to vectorize (but would like to
> because m
Dear R-helpers,
I have two problems that I don't know how to vectorize (but would like to
because my current solution is slow).
# 1.
#I have a vector x:
x <- c(3, 0, 1, 0, 2, 2, 2, 0, 4, 2)
#I want this translated into a new vector based on x,so that each element of x
#is the number of z
4 matches
Mail list logo
